In partnership with
Have you tried building your workflow automation yet?
I'm starting to explore workflow automation.
The type of automations can help grasp the fundamentals of how they work, and this usually does not include the flashy demos we see all the time.
So like any engineer, I'm starting with the basics.
The simplest possible task I could think of: fetch weather from an API and email me the temperature.
API → email.
That's it.
I tested it across 4 platforms, and gave myself a deadline for each.
Such a simple setup shouldn’t take too long, right?
Well, here are the results.
But before that, let’s see what happened in AI this week.
News News News:
TOOLS that caught my attention
1. Blink.new
YC-backed AI app builder with built-in database, auth, edge functions, and hosting. Self-corrects errors and fixes bugs without prompting.
2. Dyad
Free, open-source AI app builder that runs entirely on your machine: no cloud, no subscriptions, zero vendor lock-in.
3. Raydian
AI-first full-stack builder with integrated backend, database, auth, and hosting. Built by a NoCode agency team specifically to fix "vibe coding."
The Test
It is a simple workflow:
Input: [Location]
Process: Call OpenWeatherMap API
Output: Email with temperature
Time limit: 15 minutes per platform
Replit Agent: 6 Minutes
Prompt: "Build me an agent that checks Bengaluru weather and emails me the temperature."
Replit built it.
The interface showed it fetching the accurate weather and displaying it within 30 seconds.
Then I saw the email status: [SIMULATED]
The free tier doesn't actually send emails. It shows you what the email would look like, then does nothing. You need the paid Core plan for actual functionality.
Verdict: Fastest setup, but completely non-functional without paying.
Zapier Copilot: 5 Minutes
Copilot generated the workflow automatically: OpenWeatherMap → Gmail. Hit test, and email arrived in 30 seconds with the exact temperature.
What I liked:
Easiest email setup of any platform.
Zapier sends from its own email, no Gmail credentials needed
Actually worked on first try.
What I didn't like:
Zero customisation of email format
Stuck with Zapier's template.
"AI agent" label is misleading, this is just rebranded automation with a chat interface
Verdict: Fast and reliable if you don't need custom formatting.
n8n: 22 Minutes
Most technical of the bunch.
I had to manually configure trigger, HTTP request node, and Gmail SMTP node.
What took so long
JSON path confusion: API returns
main.tempnotmain.temperature. I spent 5 minutes just figuring out the correct path.Email credentials: Unlike Zapier, need both sender AND receiver email. Gmail requires app password, not regular password.
SMTP port settings: Port 587 needs SSL OFF. I had it ON. Took about 10 minutes of trial and error.
No hand-holding: Every error message assumed I knew what I was doing.
Once configured, the email arrived exactly as I formatted it. Complete control over every detail.
Verdict: This is built for developers who understand APIs and SMTP. If you don't know what JSON paths or SMTP ports are, you'll spend hours Googling.
Make.com: 45+ Minutes
Make is a visual builder similar to n8n.
It dragged three modules: Trigger → HTTP → Email.
And it looked the easiest to deal with. It wasn't.
30+ minutes debugging:
HTTP module failed: wrong method, then wrong headers, then wrong data mapping
Data mapping between modules broke repeatedly: would work, then suddenly wouldn't
Email configuration failed multiple times with unclear error messages
Gmail required sender credentials (not just receiver like Zapier)
Each error required Googling and trial-and-error
Eventually got the email after 45 minutes.
The problem: Visual interface creates false confidence. Dragging boxes looks easy but debugging them isn’t.
If you're non-technical, you'll hit walls fast.
If you're technical, it’d probably be faster to just write the code.
Verdict: Visual drag-and-drop positioning feels misleading if i am have to manually debug code in a no-code tool.
My Take
I initially went into this expecting "no-code" to actually mean no-code.
That's not what I found.
The tools fall into two categories:
There’s no middle ground yet.
The platforms that worked fast gave me zero flexibility.
And the platforms that gave me control required technical knowledge despite the "no-code" branding.
My takeaway: Pick based on your skill level.
I'll be using Zapier for simple stuff and n8n when I need control.
The Bigger Picture
This was the simplest possible test. If platforms struggle here, what happens when you add complexity?
Multiple API calls
Conditional logic
Error handling
Data transformation
I'm planning to test those next.
But first, I needed to know: can these platforms handle the basics? Answer: Barely. And only if you pick the right tool for your skill level.
Before picking a platform, ask yourself:
Does it work on the free tier, or just simulate?
Can I actually use it without technical knowledge?
When it breaks, can I debug it?
If the answer to any is no, save yourself the frustration.
This is just the start. I'll be building more complex automations and documenting what works (and what wastes your time).
Until next time,
Vaibhav 🤝🏻
If you read till here, you might find this interesting
#AD 1
The Headlines Traders Need Before the Bell
Tired of missing the trades that actually move?
In under five minutes, Elite Trade Club delivers the top stories, market-moving headlines, and stocks to watch — before the open.
Join 200K+ traders who start with a plan, not a scroll.
#AD 2
Help us make better ads
Did you recently see an ad for beehiiv in a newsletter? We’re running a short brand lift survey to understand what’s actually breaking through (and what’s not).
It takes about 20 seconds, the questions are super easy, and your feedback directly helps us improve how we show up in the newsletters you read and love.
If you’ve got a few moments, we’d really appreciate your insight.
